This shows you the differences between two versions of the page.

Link to this comparison view

wisar:references-on-modeling-lost-person-behavior [2014/08/11 13:32] (current)
tmburdge created
Line 1: Line 1:
 +* Maritime search: [http:// http://​www.ifremer.fr/​web-com/​stw2004/​sar/​pdf/​spaulding_ppt.pdf SAROPS]. ​ Notice the use of particle filter models, with transition probabilities being governed by currents (analogous to the influence of terrain/​vegetation) and by events/​intentions (analogous to things like trails, drainages, and decision points).
 +** I think it is an interesting parallel. Just like a disabled boat under the influence of currents, lost people can '​drift'​ also. Someone who has been lost for a while and who is exhausted is probably pretty much at the mercy of the terrain. They'​ll probably take the path of least resistance. I think the problem is being able to decide when a person will drift and when they will do things that can't be predicted with a topo map. It might be easier with boats - they probably assume the boat is damaged/​drifting if they can't reach it on the radio. I used [http://​en.wikipedia.org/​wiki/​Particle_filter Wikipedia'​s article on particle filters] in an attempt to better understand this. --Spencer Clark
 +* Wilderness search software [http://​sartechnology.ca/​sartechnology/​ST_ProgramOverview.htm Incident Commander Pro].  Notice how it only uses dispersion patterns, but doesn'​t currently use dispersion angles and distances. ​ It also doesn'​t seem to use decision points.
 +** Website: http://​sartechnology.ca/​sartechnology/​
 +** I tried the demo of this software. It's klunky and the demo is crippled, to boot. You can't even pull up the GIS interface in this demo - it brings up a picture of what the actual functionality looks like... It's hard to say what the actual program is like. --Spencer Clark
 +* Trail-based probability of area [http://​www.sarinfo.bc.ca/​Trailpoa.htm Terrain-based Approach to POA Estimation]. ​ This is by the same author as the Incident Commander Pro software. ​ I'm confused that the software doesn'​t seem to include trail-based dispersion information. ​ Someone needs to play with this software.
 +* Bayesian network and Monte Carlo methods for POA.  ''​Send email to Mike for a copy of this paper.''​ Lanny'​s preliminary work on modeling movement influenced by terrain and vegetation.
 +*I found this site that seems to be interesting. It has some available code I think we can work with - http://​sarbayes.org/​
 +** This is pretty cool. I'm surprised nobody knew about this one before? -Spencer Clark
 +*Walking Straight into Circles, by Jan Souman. I find it really intriguing -Spencer G
wisar/references-on-modeling-lost-person-behavior.txt ยท Last modified: 2014/08/11 13:32 by tmburdge
Back to top
CC Attribution-Share Alike 4.0 International
chimeric.de = chi`s home Valid CSS Driven by DokuWiki do yourself a favour and use a real browser - get firefox!! Recent changes RSS feed Valid XHTML 1.0